INDEX

Explanations

metrics indicating growth or quantities associated with success

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oni

-0.15

ione

-0.15

ene

-0.15

ÑĢÐ¾ÑģÑĤÐ¾

-0.14

lop

-0.14

IGO

-0.14

omin

-0.14

th

-0.14

Ãº

-0.14

 Deutsch

-0.14

POSITIVE LOGITS

 over

0.20

 hundreds

0.19

Hundreds

0.17

 ÏĢÎ¬Î½Ïī

0.16

 Hundreds

0.16

zens

0.15

Ð¶ÐµÐ½

0.15

 Ã¶ver

0.15

.rad

0.15

 thousands

0.15

Activations Density 0.086%

metrics indicating growth or quantities associated with success

No Comments

No Known Activations